In this study, 153 students enrolled in an undergraduate biology class for nonmajors were randomly assigned to one of four immediare corrective feedback conditions commonly used in computer-based instructional situations. In addition, the types of errors made by learners during instruction were analyzed and compared across groups. Dependent variables were achievement on a retention test, feedback study time, on-task achievement, and feedback efficiency. Results indicated that the group receiving simple knowledge-of-correct-results feedback used significantly less feedback sflAdy time and was more efficient than in any other condition. Consistent with prior studies, an adaptive design strategy overcame differences in retention that may have been observed with an instructional strategy using a fixed number of interrogatory instances. Learners who made fewer fine discrimination errors during instruction scored higher on a retention test. As expected, a stgnificantly higher number of fine discrimination errors were made on the retention test. A.n important finding of this study is that across all conditions, students used almost twice as much feedback study time when they made fine discrimination errors. (Keyword.>: science education, higher education, computerassi.>red instruction, feedback.) Developing ways to optimize a learner's acquisition and retention of intellectual skills is a central goal of those who design computer-based instruction (CBI). One of the most commonly used techniques to achieve this goal is to provide immediate corrective feedback after a learner has incorrectly responded to a question. The focus of this study was to examine the effects of four methods of immediate corrective feedback delivered within a question-based, concept and rule-learning environment. A second, perhaps more important, purpose was to probe the complex relation between types of corrective feedback and the types of errors made by learners during computer-based instruction and, later, during a retention test.